CHEMICAL EXAMINATION OF THE BLOOD. 333 



blood-serum is not a single substance, but consists of three distinct 

 albumins, which he terms a-, /?-, and ^-serum-albumin. They are 

 said to coagulate at 73 C., 77 C., and 84 C., respectively. In 

 cold-blooded animals, a-serum-albumin only is said to occur. 



More recently Oppenheimer has shown that by salting with 

 ammonium sulphate serum-albumin can be divided into two frac- 

 tions, one of which is thrown down at 66|^ per cent, saturation, 

 while the second fraction is precipitated on 82 per cent, saturation. 



Isolation. Serum-albumin is most conveniently obtained from 

 blood-serum after removal of the serum-globulin by saturation with 

 magnesium sulphate at a temperature of 30 C. The filtrate is 

 saturated with sodium sulphate or ammonium sulphate at 40 C., or 

 treated with acetic acid, so that the solution contains about 1 per 

 cent. In either case the precipitated serum-albumin is filtered off, 

 pressed between layers of filter-paper, dissolved in water (the reac- 

 tion should be neutral), and separated from the remaining salt by 

 dialysis. From its aqueous solutions it is finally obtained by 

 evaporation at a low temperature or by precipitation with alcohol, 

 which must be rapidly removed, however, as otherwise it will cause 

 coagulation of the albumin. 



Separation of the Albumins of the Blood-plasma from Each 

 Other. To isolate the fibrinogen, the plasma is treated with an 

 equal volume of a saturated solution of sodium chloride. The result- 

 ing precipitate is filtered off and purified as described. The filtrate 

 contains serum-globulin and serum-albumin. The globulin is pre- 

 cipitated by saturation with magnesium sulphate, filtered off, and 

 likewise purified. The filtrate contains only the serum-albumin, 

 which may be obtained, as just described, by saturation with sodium 

 or ammonium sulphate. 



Quantitative Estimation of the Total Albumin of the Plasma. 

 The albumins are most conveniently estimated by treating a care- 

 fully measured and weighed amount of the plasma, after neutraliza- 

 tion with acetic acid, with five times its volume of alcohol. After 

 standing for twenty-four hours the solution is boiled for several 

 minutes, and the resulting precipitate collected on a weighed filter, 

 washed with hot alcohol, then with ether, dried at 115 C., weighed, 

 and incinerated. The weight of the ash is deducted from the weight 

 of the precipitate. 



More accurate is the following method : a carefully measured 

 and weighed amount of the plasma is treated with one-half its volume 

 of a saturated solution of sodium chloride and a slight excess of 

 tannic acid. In the resulting precipitate the nitrogen is then esti- 

 mated according to KjeldahPs method. When multiplied by 6.37 

 the corresponding amount of albumin is obtained. 



To remove all albumins from the blood, Cavazzani's method may 

 be employed. To this end, 20-30 c.c. of blood are added to 200 c.c. 

 of distilled water, and treated with five or six drops of a solution 

 consisting of ten parts of acetic acid (sp. gr. 1.040) and one part of 



